I am running Red Hat 8.0, with:
cvs 1.11.2
diff (GNU diffutils) 2.8.1
I have two directories, each containing a different version of a binary file (jpeg
image), and a different version of a text file:
test-1.0/mypic.jpg
test-1.0/mytext.txt
test-1.1/mypic.jpg
test-1.1/mytext.txt
If I run:
diff -urN test-1.0 test-1.1
I get this on stdout:
Binary files test-1.0/mypic.jpg and test-1.1/mypic.jpg differ
diff -urN test-1.0/mytext.txt test-1.1/mytext.txt
<diff hunks...>
However, if I import the modules into CVS and try rdiff, things start to break:
(cd test-1.0; cvs import -I ! -W "*.jpg -k 'b'" test me V1_0)
(cd test-1.1; cvs import -I ! -W "*.jpg -k 'b'" test me V1_1)
cvs -q rdiff -u -rV1_0 -rV1_1 test
Now I get this (I've prepended stdout: and stderr: to each line):
stdout: Index: test/mypic.jpg
stderr: cvs rdiff: failed to read diff file header /mtp/cvsxGIzHz for mypic.jpg,v: end
of file
stdout: Index: test/mytext.txt
diff -u test/mytext.txt:1.1.1.1 test/mytext.txt:1.1.1.2
<diff hunks...>
The output of this command is 1 (failure).
Shouldn't rdiff act like diff, and just print out that the binary files differ?
What exactly is the point of the cvswrappers option during import (-W)... it doesn't
seem to matter with rdiff. Is this just an rdiff quirk, or am I missing something?
Tony Clayton
_______________________________________________
Bug-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-cvs
